Neural differentiation involves a complex train of events driven by intrinsic programming of the cell, cell-cell interactions, and epigenetic phenomena within the microenvironment. The goal of project 2 is to examine molecular events governing granule neuron development and the role(s) this cell plays in the establishment of cerebellar cytoarchitecture. To date, there are no widely available markers for either cerebellar granule neurons or granule precursor cells. Therefore, MN20 has been recently cloned in the PI's laboratory from an early postnatal granule cell cDNA library by virtue of its cross hybridization with a conserved region of Drosophila gene, called Notch, which acts in the determination of cell fate during neurogenesis. This region contains 36 repeated units bearing homology with epidermal growth factor (EGF). MN20 is heavily expressed in mitotic granule neuron precursors in the external granule layer (EGL) and is virtually turned off in the adult cerebellum. Thus, the MN20 gene presents an opportunity for the examination of molecular and cellular interactions in the developing cerebellum. MN20 will be used as a molecular marker of granule neuron differentiation with which to examine transcriptional regulatory mechanisms operating during the neurodevelopment of this cell. In order to examine the role of these cells in cerebellar development, the MN20 promoter will be used to produce transgenic mouse models in which granule neuron populations are deleted. the approach will include the examination of the: 1) temporal and spatial distribution of MN20 gene expression; 2) structure of the MN20 promoter from genomic DNA; 3) regulation of MN20 gene transcription. This will serve as a model for aspects of transcription regulation during granule cell development; 4) functional role of granule neurons in the formation of the cerebellum using transgenic models. (a) MN20-lacZ transgenes will be used to determine the extent of the MN20 promoter required to obtain expression most restricted to granule neuron precursors. Predicated on information obtained in 1 and 4a, MN20 promoter, fused to coding sequence for diphtheria toxin, will be used to deplete early granule neuron populations. (b) Alternatively, the consequences of MN20 expression in the developing cerebellum will be investigated using a retroviral-based construct to provide a continuous intracellular source of antisense MN20 RNA, thereby reducing the levels of protein produced. The antisense approach will be used sequentially in 3 paradigms: i) cerebellar cells in primary culture, ii) infection of EGL cells by intracranial injection into early postnatal mice, and iii) if successful reduction in MN20 expression is demonstrated in (i) and (ii), MN20 antisense transgenic mice will be produced. In conjunction with the other projects in this program, these studies will develop tools with which to examine the cellular and molecular interrelationships of cell types in the formation of the cerebellum.